US technology stocks experienced a significant shift in investor positioning on Wednesday, as the Democratic Party’s election victory raised concerns about potential restrictions on new data centre developments. Investors moved toward established cloud computing companies while reducing exposure to semiconductor manufacturers and AI infrastructure developers.
Amazon (NASDAQ:AMZN) climbed 6.1%, Microsoft (NASDAQ:MSFT) advanced 6.7%, Alphabet (NASDAQ:GOOG) gained 3.2% and Meta Platforms (NASDAQ:META) rose 1.4%.
The contrasting performance highlighted a growing distinction between companies operating existing AI infrastructure and those relying on future capacity expansion.
Goldman Sachs Reports Strong Demand for Hyperscalers
Goldman Sachs identified substantial buying interest in major cloud providers from both long-only investment funds and hedge funds.
“Our US desk is seeing a clear rotation into Hyperscalers from both LOs and HFs,” the bank’s trading desk reported.
Goldman’s technology specialists suggested that the election results had changed investor expectations regarding the development of new computing infrastructure.
“Investors viewed the Democratic win as potentially making new builds harder and existing, power-secured compute more valuable,” they explained.
Potential changes to permitting requirements, water consumption rules, electricity access and tax incentives could make new data centre construction more expensive and time-consuming.
Existing hyperscalers with operational facilities and secured power supplies could consequently benefit from reduced competition and increased pricing power.
“Scarcer powered capacity could lift prices and strengthen existing operators’ competitive advantage, while leaving future-capacity providers behind,” Goldman’s specialists added.
CoreWeave and Nebius Fall as Infrastructure Stocks Weaken
While established cloud providers attracted buying interest, companies focused on developing additional AI computing capacity experienced substantial losses.
CoreWeave (NASDAQ:CRWV) dropped 9.9%, while Nebius (NASDAQ:NBIS) declined 6.7%.
Other neocloud providers and data centre construction companies also weakened, alongside semiconductor and memory manufacturers such as Micron Technology (NASDAQ:MU).
The declines reflected concerns that regulatory restrictions and electricity shortages could delay the deployment of new computing infrastructure.
Data Centre Capacity Forecasts Face Political Risks
The market rotation was reinforced by warnings about potential constraints on future data centre construction.
Investor Brad Gerstner suggested during an appearance on the All-In podcast that regulatory obstacles following the Democratic victory could prevent between 10 and 15 gigawatts of expected data centre capacity from becoming operational in 2027.
Goldman noted that these estimates were already below some more optimistic industry projections.
In September, Gerstner had suggested that approximately 25 gigawatts of US capacity represented a realistic development target.
Separately, research firm SemiAnalysis identified 16.7 gigawatts of planned Texas data centre capacity for 2028 as potentially vulnerable.
“The main point is that the election now adds political risk on top of already bullish street forecasts well above 30GW,” Goldman stated.
Semiconductor Companies Face Potential Inventory Risks
Restrictions on data centre development could create additional challenges for semiconductor manufacturers if chip deliveries outpace the availability of operational computing facilities.
Goldman highlighted concerns that delays in electricity connections and construction could lead to excess semiconductor inventory.
“There is a risk now is that the semis narrative flips to inventory correction if power/sites lag chip deliveries,” the bank noted, summarising sceptical market views.
According to a Financial Times report cited in the analysis, Amazon Web Services has been receiving approximately $3 billion worth of Nvidia GB300 server racks each month, with some equipment reportedly being stored because certain facilities lack sufficient electricity capacity.
The situation has drawn comparisons with the semiconductor inventory correction of 2022 and 2023.
Higher Computing Costs Could Challenge the AI Investment Case
Beyond immediate market movements, Goldman raised questions about the longer-term consequences of restricted computing capacity.
Sustained advances in artificial intelligence depend partly on declining computing costs, which allow businesses to deploy increasingly sophisticated systems at greater scale.
If tighter regulations permanently increase the cost of developing new infrastructure, rather than merely delaying construction, the economic outlook for AI adoption could weaken.
Higher computing expenses could also reduce the competitive pressure AI places on traditional software and media companies.
The recent rotation therefore reflects more than a temporary shift between technology sectors. Investors are increasingly assessing which companies could benefit from scarce computing capacity and which face greater risks from infrastructure constraints.